Alignment Help

[[Da...]]

I have been asked to make a program for the attached part. It is very small (OD is .486 inch)

The only way I could think to fixture it is by mounting it with a screw onto a milled down scrap part. (see photo)

But now I am struggling with how to align it for the following reasons:

No flat spot on part to use as a Z zero

I have never tried using a torus as the X and Y zero. Is this even allowed, and if so, are there any special considerations when doing this?

I don't know what kind of strategy to use on a torus. Just a bunch of random points?

I considered using the shoulder of the fixture as the Z zero and the cylinder on the fixture below the part as the  and Y Zero. I believe I would then have to make a secondary alignment using features of the actual part in order to get the angle of the indent features, but that takes me back to the original problem. 

I was going to clock it to the dimple, but it is very small and I don't know how to probe it. Is there a better way to clock it? If I did want to clock the dimple, how would I do that when the probe hardly fits inside? Is there a way to have the probe find the center of the dimple like it does when you are setting the probe rack location?

 

Contura with XXT. I only have base Calypso. No curve or freeform. 

[[DW...]]

Please sign in to view this username.

 What are you trying to measure? What are the tolerances for those characteristics? I would argue the CMM is not the best way to inspect this piece. 

[[RP...]]

What probes do you have available?

I'd use a 1.0mm star probe to extract all the features.  clocked with the dimple.  add in circle scans and recall into lines for the funky angled mill areas.  secondary alignments for each of them.  or space points all around and calculate deviations.

[[Da...]]

Please sign in to view this quote.

The angles radii both directions on the slanted indents, as well as the angles of the slanted indents to the top of the part. 

I agree the CMM might not work out well, but I was asked to try. 

[[DW...]]

Please sign in to view this username.

 Yea the juice is probably not going to be worth the squeeze with the CMM. This is a light/vision system job, maybe with some professional putty.  

[[Da...]]

Please sign in to view this quote.

We also have a Keyence IM but its an older one and I haven't had much luck with small parts like this in the past so I don't know if that will be any better. 

[[DW...]]

Please sign in to view this username.

 Is this just a one off verification or an ongoing production check? Might be worth it to just outsource it to a third party inspection laboratory if you have the budget. 

[[Da...]]

Please sign in to view this quote.

Can you please walk me thru what strategy I would use for the dimple so that I can clock to it? I do have a 1mm star probe and regular probe as well as some larger straight probes. I want to probe the dimple and have it find the center like how it does when you locate the probe rack, but I don't know how to do that. The dimple is much to small to do the strategies I would use if it were bigger. 

[[Jo...]]

You can identify features too small to probe by scanning over the area and recalling the lowest -Z point from the scan as the location of your dimple. 

This strategy is taught in the Zeiss expert class specifically as a way to locate a clocking feature with an unknown orientation

[[Da...]]

Please sign in to view this quote.

Great tip, thank you!

How do I recall the lowest -Z point?

[[Zo...]]

Please sign in to view this quote.

use minimum or maximum coordinates

 

 

 

[[Zo...]]

btw, using torus is a lot less scarier than it seems

you can use a combination of vertical (small diameter) and horizontal circle (in your case in four segments) using your star probe

specially when you have the cad model

[[Da...]]

Please sign in to view this quote.

This is my first time trying to probe a torus, and I agree it seems pretty straightforward...the only issue is that it is so small that no matter how small of paths I set, it the probe keeps going over the edges. 

[[Ke...]]

Were you able to extract the feature as a single Torus, from the CAD model?

The CAD lines in the middle of the feature would lead me to believe that we are looking at two seperate Torus features and a thin Cylinder that seperates them.

What type of measurements are you trying to take from it?

Edited 4 hours ago

[[Je...]]

Please sign in to view this quote.

I respectfully disagree with you on this one,

Please sign in to view this username.

.  We fixture and measure parts of that size and shape all day long.

It's all a matter of scale.

The helix geometry on that part lend itself to either tactile or DotScan non-contact.  Vision may be helpful, but it could also introduce projection error.

[[Je...]]

Please sign in to view this quote.

It's okay to overscan the paths and then limit the evaluation to only the portions inside the area of interest.

Also, keep an eye out for whether Calypso extracts the Torus correctly as either an internal or external feature.  It's been prone to invert the two on certain versions.